Joel, the original release of the nx-amp used the KSC3503 and KSA1381 but I later changed these to KSC2690 and KSA1220. However, these are now discontinued so the design has reverted back to the 3503/1381. The original design was fully characterized with the 3503/1381 which used two adjacent gain grades. If you can, select the 3503/1381 for the closest gain match - but it’s not critical, your amp will work fine.
Either of the two BC556 devices will work (One has the preformed leads and the other the straight leads).
Apologies for the tardy reply.
Either of the two BC556 devices will work (One has the preformed leads and the other the straight leads).
Apologies for the tardy reply.
Thanks Bonsai. I thought ABU was gain grade A , A being in front of B. Either way, I've secured BTF for both 546 and 556,
cheers!
cheers!
I enjoyed reading the design goals for Ovation SX. I have future plans for a speaker build for a friend, tri-amped with class A amps. Looking at 40-50w rms per channel.. The SX has some hefty output devices.. Is 40-50w possible with larger heatsinks and higher voltage transformers? Or not possible? Thanks in advance.. Either way, I'm prepared to order some pcbs and try it out at 15w
Joel, I’d recommend the kx2 rather than the sx-Amp for this application. You can easily switch between class A, AAB and AB. The kx2 will also do supply rails up to 35V giving ~80W into 8 Ohms in class AAB. The distortion is also very low in all modes.
Yes, I know, I'm building the nx amplifier too slowly and I continue to ask questions ... Please be patient.
I started measuring (hoping that the measurement tool is enough accurate) the BC546B and the BC556B, and I discovered that it will be difficult to match their characteristics.
The BC546 (NPN) have values consistently around hFE=300, Ic=2.8mA.
The BC556 (PNP) can be divided into two groups: hFE=360, Ic=2.2mA and hFE=430, Ic=6.1mA.
Looking at the datasheets this dispersion seems normal.
Am I right?
What can I do?
I started measuring (hoping that the measurement tool is enough accurate) the BC546B and the BC556B, and I discovered that it will be difficult to match their characteristics.
The BC546 (NPN) have values consistently around hFE=300, Ic=2.8mA.
The BC556 (PNP) can be divided into two groups: hFE=360, Ic=2.2mA and hFE=430, Ic=6.1mA.
Looking at the datasheets this dispersion seems normal.
Am I right?
What can I do?
teodorom,
just use the closest match - your amp will still work perfectly. You only need to match to squeeze a bit lower distortion.
🙂
just use the closest match - your amp will still work perfectly. You only need to match to squeeze a bit lower distortion.
🙂
OK, thank you. I can match them at 10%. I guess that the SPICE models too have similar differences.
Finished measuring the BC547C and BC557C! Again I can get another 10% match.
On the other side the SPICE models (that I have checked with the specs) show the same differences.
I shall use those together with the BC546B and BC556B.
Found on TME that has a good stock of DIOTEC transistors.
Hope it helps.
On the other side the SPICE models (that I have checked with the specs) show the same differences.
I shall use those together with the BC546B and BC556B.
Found on TME that has a good stock of DIOTEC transistors.
Hope it helps.
I am making a new case for my SX amp and the existing case will be up for sale - it might help a new builder. It has large heatsinks and only runs to about 40C in the centre. It has a hole in the front panel for a mains switch and two holes for "on" LEDs. The front panel is tapped to hold the little pcbs for these, which I will include. The heatsinks are tapped for the amp mounting and transistor fixing. I will include these screws and thermal pads so that completing a build is easy. Unfortunately there is a scratch on the top panel. The case weighs 8.14kg (18lb) so the shipping cost might put off people outside the UK.
Send me a pm if interested.
Send me a pm if interested.
Thought I should mention, my 2 latest NX PSU boards don't have silk print on the back to identify the components. I will use the older board for component identification.
Last edited:
Joel
sorry to hear this - I will send Stanton an email about your problem. If you need any help with assembly, please ping me.
sorry to hear this - I will send Stanton an email about your problem. If you need any help with assembly, please ping me.